Sound processing with increased noise suppression

Abstract

A method for processing sound that includes, generating one or more noise component estimates relating to an electrical representation of the sound and generating an associated confidence measure for the one or more noise component estimates. The method further comprises processing, based on the confidence measure, the sound.

What is claimed is: 
     
       1. A method, comprising:
 receiving at least one sound signal at a hearing device having an array of electrodes; 
 determining an instantaneous signal-to-noise ratio (SNR) of the at least one sound signal; 
 applying a gain that varies with the instantaneous SNR to the at least one sound signal to generate a noise-reduced signal from the at least one sound signal, wherein when the instantaneous SNR reaches a gain threshold of approximately 5 decibels (dB), applying the gain includes applying a gain of less than approximately −5 dB to the at least one sound signal to generate the noise-reduced signal from the at least one sound signal; and 
 generating a control signal for controlling stimulation of at least one electrode of the array of electrodes using the noise-reduced signal. 
 
     
     
       2. The method of  claim 1 , wherein applying a gain of less than approximately −5 dB to the at least one sound signal to generate the noise-reduced signal from the at least one sound signal, comprises:
 applying a gain of less than approximately −6.5 dB to the at least one sound signal. 
 
     
     
       3. The method of  claim 1 , wherein applying a gain of less than approximately −5 dB to the at least one sound signal to generate the noise-reduced signal from the at least one sound signal, comprises:
 applying a Wiener function to the at least one sound signal. 
 
     
     
       4. The method of  claim 1 , further comprising:
 selecting the gain to be applied to the at least one sound signal to generate the noise-reduced signal from a gain curve that varies with the instantaneous SNR of the at least one sound signal. 
 
     
     
       5. The method of  claim 1 , wherein the applied gain at least partially cancels a noise component within the at least one sound signal and at least partially cancels a speech component in the at least one sound signal. 
     
     
       6. The method of  claim 1 , wherein the applied gain is selected from a smooth gain curve represented by a parametric Wiener function. 
     
     
       7. The method of  claim 1 , further comprising:
 splitting the at least one sound signal into a plurality of spectral components, 
 wherein determining an instantaneous SNR of the at least one sound signal comprises determining an instantaneous SNR of at least one of the plurality of spectral components. 
 
     
     
       8. The method of  claim 7 , wherein applying the gain of less than approximately −5 dB to the at least one sound signal to generate the noise-reduced signal from the at least one sound signal comprises:
 applying the gain of less than approximately −5 dB to the at least one of the plurality of spectral components. 
 
     
     
       9. The method of  claim 7 , wherein applying the gain of less than approximately −5 dB to the at least one sound signal to generate the noise-reduced signal from the at least one sound signal comprises:
 applying the gain in accordance with a gain function having any section which lies between a parametric Wiener gain function parameter values of α=0.12 and β=20, and a parametric Wiener gain function parameter values of α=1 and β=20, over a range of instantaneous SNRs between approximately −5 dB and approximately 20 dB. 
 
     
     
       10. The method of  claim 1 , further comprising:
 when the instantaneous SNR reaches approximately 10 dB, applying a gain of less than approximately −2 dB to the at least one sound signal to generate the noise-reduced signal from the at least one sound signal. 
 
     
     
       11. A method, comprising:
 receiving at least one sound signal at a hearing device having an array of electrodes; 
 determining an instantaneous signal-to-noise ratio (SNR) of the at least one sound signal; 
 applying a gain that varies with the instantaneous SNR to the at least one sound signal to generate a noise-reduced signal from the at least one sound signal, wherein when the instantaneous SNR reaches a gain threshold of approximately 10 decibels (dB), applying the gain includes applying a gain of less than approximately −15 dB to the at least one sound signal to generate the noise-reduced signal from the at least one sound signal; and 
 generating a control signal for controlling stimulation of at least one electrode of the array of electrodes using the noise-reduced signal. 
 
     
     
       12. The method of  claim 11 , wherein applying a gain of less than approximately −15 dB to the at least one sound signal to generate the noise-reduced signal from the at least one sound signal, comprises:
 applying a Wiener function to the at least one sound signal. 
 
     
     
       13. The method of  claim 11 , further comprising:
 selecting the gain to be applied to the at least one sound signal to generate the noise-reduced signal from a gain curve that varies with the instantaneous SNR of the at least one sound signal. 
 
     
     
       14. The method of  claim 11 , wherein the applied gain is selected from a smooth gain curve represented by a parametric Wiener function. 
     
     
       15. The method of  claim 11 , further comprising:
 splitting the at least one sound signal into a plurality of spectral components, 
 wherein determining an instantaneous SNR of the at least one sound signal comprises determining an instantaneous SNR of at least one of the plurality of spectral components. 
 
     
     
       16. The method of  claim 15 , wherein applying the gain of less than approximately −15 dB to the at least one sound signal to generate the noise-reduced signal from the at least one sound signal comprises:
 applying the gain of less than approximately −15 dB to the at least one of the plurality of spectral components. 
 
     
     
       17. The method of  claim 11 , wherein applying the gain of less than approximately −15 dB to the at least one sound signal to generate the noise-reduced signal from the at least one sound signal comprises:
 applying the gain in accordance with a gain function having any section which lies between a parametric Wiener gain function parameter values of α=0.12 and β=20, and a parametric Wiener gain function parameter values of α=1 and β=20, over a range of instantaneous SNRs between approximately −5 dB and approximately 20 dB. 
 
     
     
       18. A hearing device, comprising:
 one or more inputs configured to receive at least one sound signal; and 
 at least one processor configured to:
 determine an instantaneous signal-to-noise ratio (SNR) of the at least one sound signal; 
 generate a noise-reduced signal from the at least one sound signal by applying a gain that varies with the instantaneous SNR to at least one portion of the at least one sound signal, wherein when the instantaneous SNR reaches a gain threshold of approximately 5 decibels (dB), the noise-reduced signal is generated from the at least one sound signal by applying a gain of less than approximately −6.5 dB to the at least one portion of the at least one sound signal; and 
 generate, based on the noise-reduced signal, at least one control signal for controlling one or more electrical stimulation signals for delivery to a recipient of the hearing device. 
 
 
     
     
       19. The hearing device of  claim 18 , further comprising:
 an array of electrodes configured to be implanted in the recipient; and 
 a stimulator unit configured to generate, based on the at least one control signal, one or more stimulation signals for delivery to the recipient via one or more of the electrodes. 
 
     
     
       20. The hearing device of  claim 18 , wherein the at least one processor is configured to:
 determine the gain to be applied to the at least one portion of the at least one sound signal to generate the noise-reduced signal from again curve that varies with the instantaneous SNR of the at least one sound signal, 
 wherein the gain curve from which the applied gain is selected is a smooth gain curve represented by a parametric Wiener function. 
 
     
     
       21. The hearing device of  claim 18 , wherein the at least one processor is configured to:
 apply the gain in accordance with a gain function having any section which lies between a parametric Wiener gain function parameter values of α=0.12 and β=20, and a parametric Wiener gain function parameter values of α=1 and β=20, over a range of instantaneous SNRs between approximately −5 dB and approximately 20 dB.